Bahir Dar Institute of Technology( BiT.

)
Faculty of Mechanical and Industrial Engineering
Industrial Engineering Program

INDUSTRIAL MANAGEMENT AND ENGINEERING

ECONOMICS

Chapter 3: Forecasting

March, 2025
1
Outline

▪Forecasting Framework
▪Characteristics of Demand Forecasting
▪Useful Forecasting Models
▪Forecasting Errors

2
 A Forecasting Framework
 Forecasting is the art and science of predicting future events.
 Forecasting demand is a critical aspect of operations management,
as it helps organizations align their production capabilities with
customer needs (Demand may differ sales).
 Demand management focus on strategies and processes to
coordinate and control demand ensuring that the production system
operates efficiently.
 Demand can be:
 Dependent: demand for a product caused by the demand for
other product (a product linked to demand of other)
✓ Components required for manufacturing a finished product (e.g.,
tires for cars).
 Independent: occurs independently of demand from any other
product (can not be derived directly from that of other product)
✓ Consumer goods sold directly to customers (e.g., electronics,
clothing). 3
 A Forecasting Framework
 Difference between forecasting, prediction and planning.
• Forecasting: what we think will happen
• Planning: what we think should happen

❑ Forecasting is a process of estimating a future event by casting

forward past data. The past data are

➢Systematically combined in a predetermined way to

obtain the estimate of the future.

4
 Cont.…
❑Prediction is a process of estimating a future event based on
subjective considerations other than just past data;

➢ These subjective considerations need not be combined

in a predetermined way.

❑Planning is the process of thinking about the future course of

action which is required to achieve a specific goal.

✓It explains about what course of action is required to be taken,

when is the right time, by whom and where. Also, it explains the
best scenario, the worst scenario, and the most expected case etc.
5
 Cont.…
 Forecasting is an input to all business planning and control
• Marketing uses for planning product, pricing, promotion and
placement
• Finance uses for financial planning
• Forecasting for operation decision
 Forecasting application in various decision areas of
operations (capacity planning, inventory management,
others)

6
 Cont.…
• Importance of Forecasting
➢ To plan for the future by reducing uncertainty.
➢ To anticipate and manage change.
➢ To increase communication and integration of
planning teams.
➢ To anticipate inventory and capacity demands and
manage lead times.
➢ To project costs of operations into budgeting
processes.
➢ To improve competitiveness and productivity
through decreased costs and improved delivery and
responsiveness to customer needs.

7
 Forecasting: A Decision Making Process

▪The essential problem of management is to transform a company’s strategic

objectives into decision and action.

▪The increasingly unstable dynamics of the business environment highlight the

essential role of forecasting in the decision-making process

▪Forecasting Horizon

▪Short term (0 to 3 months): for inventory management and scheduling

▪Medium term (3 month to 2 years): for production planning, purchasing

and distribution

▪Long term (2 years and more): for capacity planning, facility location and
strategic planning
8
 Forecasting and operation system
Information on most recent demand and
production

Demand forecast for

operations

Planning the system (design) Controlling the system

• Product design Scheduling the system •Production control
• Process design • Aggregate production •Inventory control
• Equipment investment and planning •Labor control
replacement • Operation scheduling •Cost control
• Capacity planning

Output of goods and

9 services
 Characteristics of demand over time
• Time series analysis
Plot the demand data on a time scale, study the plot and look
for consistent shape or pattern. The time series of demand
might have the following pattern
• Constant
• Trends
• Seasonal and cyclical pattern
• Random variation (cause by chance of event)
• Some combinations of these patterns
• Conditions
• Low noise: most the points lies around /very close to the
pattern
• High noise: many points lies relatively far away from the
pattern
10
Time Series Components

11
Cont.…

12
 Components of Demand
Trend Component
✓ Persistent, overall upward or
downward pattern
✓Factors such as population growth,
technological advancements, changes in
consumer preferences, and cultural
shifts can influence trends.
✓Trends can last for several years,
indicating sustained changes in demand.
◆Seasonal Component
✓Regular pattern of up and down
fluctuations
✓Influenced by factors such as weather
conditions, cultural customs, and
holiday shopping behaviors.
◆Occurs within a single year

13
Components of Demand
Cyclical Component *
◆ Repeating up and down
movements
◆ Affected by business cycle,
political, and economic factors
◆ Multiple years duration
Random Component 0 5 10 15 20

◆ Erratic, unsystematic, ‘residual’

fluctuations
◆ Changed by due to random
variation or unforeseen events
◆ Random fluctuations are typically
short-lived and nonrepeating
M T W T F
14
 Useful Forecasting Model
Techniques
• Qualitative and judgment method: based on expert estimates and
opinions other than numerical data's
• Naïve: Assumes demand in next period is the same as demand in
most recent period (e.g., If January sales were 68, then February
sales will be 68). Sometimes cost effective and efficient . Can serve
as a good starting point for more complex models.
• Time series (quantitative/Extrapolative model): based on data
related to past demand can be used to predict future demand.
Extrapolative model: extending existing patterns into the future
• Causal relationship (quantitative) or Explanatory model: uses
linear regression, to understand the relationship between demand
and one or more independent variables. 15
 ‘Qualitative’ Forecasting Methods
• Based upon managerial judgment when there is a lack of data or
the data were not reliable or relevant. No specific model is used
but qualitative.

• When new product is introduced

• Major methods:
Grass root
• Builds the forecast by adding successively from the bottom,
the assumption is that the person closest to the customer or
end user of the product knows its future need best (front
line operators data)
16
 ‘Qualitative’ Forecasting Methods
Delphi Technique
• It is a panel of group of experts with different level of expertise
(variety of knowledgeable people) and answer questionnaires
and summarized given back to the entire group with new set of
questions
• Delphi conceals the identity of individuals participating in the
study
Market Surveys (research)
• Panel, questionnaire, market test by prototypes without full
production
Life-cycles (historical) Analogy
• In forecasting new products, where an existing product or
generic product could be used as a model
Informed Judgment
✓Involves gathering insights from a group of individuals with
relevant experience and knowledge.
✓This group can include marketers, product developers, and industry
experts.
17
 ‘Quantitative’ Forecasting Methods
Time-Series Forecasting
•This approach is particularly useful for identifying trends, seasonal patterns, and
cyclical behaviors in data over time.
•Try to predict the future based on the past data
•To select forecasting model: Time horizon, Data availability, Accuracy required,
size of forecasting budget, qualified personnel and decomposition of data’s
Common Forecasting Methods
•Simple Moving Average
•Weighted Moving Averages
•Exponential Smoothing
•Regression Analysis
9
 Cont.…
Naive:
• The forecast is equal to the actual value observed during the last
period – good for level patterns
Simple Mean:
Ft +1 = At
• The average of all available data - good for level patterns
Ft +1 =  A t / n
Simple Moving Average
•Assumes no trend, seasonal or cyclical components.
•Simple Moving Average: combines demand data from several of the
most recent periods; their average being the forecast for next period.
•As general rule: the longer the averaging period, the slower response
to demand change
𝐷𝑡 + 𝐷𝑡 −1 + … + 𝐷𝑡 −𝑁+1
𝐴𝑡 =
𝑁
19 𝐹𝑡+1 = 𝐴𝑡
 Simple Moving Average

• Forecast Ft is average of n previous observations or actuals Dt

:
1
Ft +1 = ( Dt + Dt −1 +  + Dt +1− n )
n
1 t
Ft +1 =  Di
n i = t +1 − n
• Note that the n past observations are equally weighted.

• Issues with moving average forecasts:

• All n past observations treated equally;
• Observations older than n are not included at all;
• Requires that n past observations be retained;
 Simple Moving Average
• Include n most recent
observations • A smaller N makes the forecast
• Weight equally more responsive.
• Ignore older observations • A larger N makes the forecast
more stable. stability helps to
reduce the impact of short-term
weight variability and noise in the data.

1/n

n ... 3 2 1
today
21
 Simple Moving Average
Compute three period moving average (number of periods
is the decision of the forecaster)
Period Actual Demand Forecast
1 10
2 18
3 29
4 19

(10+18+29)/3 = 19
Period 5 will be (18+29+actual for period 4)/3

22
 Simple Moving Average Example

Actual 3-Month
Month Shed Sales Moving Average

January 10
February 12
March 13
April 16 (10 + 12 + 13)/3 = 11 2/3
May 19 (12 + 13 + 16)/3 = 13 2/3
June 23 (13 + 16 + 19)/3 = 16
July 26 (16 + 19 + 23)/3 = 19 1/3
 Simple Moving Average Example

• Example: forecasting sales

➢Actual sells of bottled spring water

What will
the sales be
for July?
Simple Moving Average Example
 Weighted Moving Average

• Weighted Moving Average: wants to use the moving

average but does not want to have all n periods equally
weighted. This makes responsive:

26
 Cont.…
Example:
Weights Applied Period
Weighted
3 Last month
Moving 2 Two months ago
Average 1 Three months ago
6 Sum of weights
Actual 3-Month Weighted
Month Shed Sales Moving Average

January 10
February 12
March 13
April 16 [(3 x 13) + (2 x 12) + (10)]/6 = 121/6
May 19 [(3 x 16) + (2 x 13) + (12)]/6 = 141/3
June 23 [(3 x 19) + (2 x 16) + (13)]/6 = 17
July 26 [(3 x 23) + (2 x 19) + (16)]/6 = 201/2
 Exponential Smoothing: Concept
• Include all past observations

• Form of weighted moving average

• Weight recent observations much more heavily than very old observations:

weight

Decreasing weight given 

to older observations
(1 − )
(1 − )2
(1 − ) 3

today
28
 Simple Exponential Smoothing
• The forecast: Denotes the
𝑭𝒕+𝟏 = 𝑭𝒕 +∝ (𝑫𝒕 − 𝑭𝒕 ); Smoothing
constant importance of
𝑭𝒕+𝟏 =∝ 𝑫𝒕 + (𝟏−∝)𝑭𝒕 ; alpha α the past error
𝑭𝒕 =∝ 𝑫𝒕−𝟏 + (𝟏−∝)𝑭𝒕−𝟏 ;
𝟎 𝟏 𝟐 𝟑
𝑭𝒕+𝟏 =∝ 𝟏−∝ 𝑫𝒕 +∝ 𝟏−∝ 𝑫𝒕−𝟏 +∝ 𝟏−∝ 𝑫𝒕−𝟐 + 𝟏−∝ 𝑭𝒕−𝟐 ;

F=forecast of demand (both this period and next)

D = actual demand (this period)
t = time period
The value of the smoothing constant () is a choice. It
determines how much the calculation smooths out the random
variations. Its value can be set between zero (0) and one (1).
Normally it is in the 0.1 to 0.3 range. 29
 Exponential Smoothing-calculation
➢If Ft+1 is to be very responsive to recent demand, choose the
large value of 
• No trend, cyclical or seasonal components.
• The most recent occurrences are more indicative of the future
than in the more distant past
• Facts:
• September forecast for sales was 15
• September actual sales were 13
• Alpha ( α) is 0.2
• What is the forecast for October?
• Calculation
• October Forecast = September forecast + α(September
actual-September forecast)
=15+0.2(13-15)=15+0.2(-2)=15-0.4=14.6
30
 Exponential Smoothing Example

Predicted demand (t-1)= 142 Ford Mustangs

Actual demand (t-1) =153
Smoothing constant a = .20

New forecast (t) = 142 + .2(153 – 142)

 Exponential Smoothing Example

Predicted demand = 142 Ford Mustangs

Actual demand = 153
Smoothing constant a = .20

New forecast = 142 + .2(153 – 142)

= 142 + 2.2
= 144.2 ≈ 144 cars
 Impact of Different 

225 –

Actual  = .5
200 – demand
Demand

175 –

 = .1
150 – | | | | | | | | |
1 2 3 4 5 6 7 8 9
Quarter
© 2011 Pearson Education, Inc.
publishing as Prentice Hall
 ◆ Chose high values of 
Impact of Different  when underlying average
is likely to change
◆ Choose low values of 
225 – when underlying average
is stable
Actual  = .5
200 – demand
Demand

175 –

 = .1
150 – | | | | | | | | |
1 2 3 4 5 6 7 8 9
Quarter
© 2011 Pearson Education, Inc.
publishing as Prentice Hall
Choosing 
➢The objective is to obtain the most accurate
forecast no matter the technique
➢We generally do this by selecting the model that
gives us the lowest forecast error\
➢Forecast error = Actual demand - Forecast value
= At - Ft

35
 Forecast Errors
Error Estimation might be used

 To monitor erratic demand observations and outliers (Perhaps

may be rejected from data)

 To determine when the forecasting method is no tracking actual

demand and needs to be reset

 To determine the parameter values (N and α) that provide the

forecasting with least error

 To set safety stock or safety capacity and ensure against stock out

36
 Forecast Errors
• Forecasts are never perfect.

• Need to know how much we should rely on our chosen

forecasting method.

• Measuring forecast error:

• Note that over-forecasts = negative errors and under-

forecasts = positive errors.

37
 Forecast Errors measures

 Cumulative Sum of Forecast Error (CFE) and Mean Forecast

Error (MFE)

 Mean Square Error (MSE)

 Mean Absolute Deviation (MAD)—measure of deviation in

units.

 Mean Absolute Percentage Error (MAPE)

 Tracking Signal (TS)—relative measure of bias

 Forecast error for Period t is et

◦ et=Actual demand (Dt)- Forecast (Ft)
11-38
 Forecasting Performance

How good is the forecast?

• Mean Forecast Error (MFE or Bias): Measures average
deviation of forecast from actuals.
• Mean Absolute Deviation (MAD): Measures average
absolute deviation of forecast from actuals.
• Mean Absolute Percentage Error (MAPE): Measures
absolute error as a percentage of the forecast.
• Standard Squared Error (MSE): Measures variance of
forecast error
39
 Mean Absolute Deviation (MAD)
n
1
MAD =
n

t =1
D t − Ft

• Measures absolute error

• Positive and negative errors thus do not cancel out (as with MFE)

• Want MAD to be as small as possible

• No way to know if MAD error is large or small in relation to the

actual data

40
Forecast Errors: Formulas

If TS > 4 or < -4, investigate!

 Tracking Signal
▪ Analogous to control charts in quality control, viz. if there is
no bias, its values should fluctuate around zero.
▪ Is a relative measure, i.e. the numbers mean the same for
any forecast.
Signal exceeding limit
Tracking signal
Upper control limit
+

0 MADs Acceptable
range

–
Lower control limit

Time 11-42
 Tracking Signal Example

Cumulative
Absolute Absolute
Actual Forecast Forecast Forecast
Qtr Demand Demand Error CSFE Error Error MAD

1 90 100 -10 -10 10 10 10.0

2 95 100 -5 -15 5 15 7.5
3 115 100 +15 0 15 30 10.0
4 100 110 -10 -10 10 40 10.0
5 125 110 +15 +5 15 55 11.0
6 140 110 +30 +35 30 85 14.2
 Tracking Signal Example

Cumulative
Absolute Absolute
TrackingForecast
Actual Forecast Forecast
Qtr Signal Demand
Demand Error CSFE Error Error MAD
(CSFE/MAD)
1 90 100 -10 -10 10 10 10.0
-10/10 = -1
2 95 = -2100
-15/7.5 -5 -15 5 15 7.5
3 115
0/10 = 0 100 +15 0 15 30 10.0
4 -10/10
100 = -1110 -10 -10 10 40 10.0
+5/11 = +0.5
5 125 = +2.5
+35/14.2 110 +15 +5 15 55 11.0
6 140 110 +30 +35 30 85 14.2

The variation of the tracking signal between -2.0 and +2.5

is within acceptable limits
 Example

• During the past 8 quarters, the Port of Djibouti has unloaded

large quantities of grain from ships. The Port’s Operations
Manager wants to test the forecasting method exponential
smoothing to see how well this method works in predicting
tonnage unloaded. He guesses that the forecast of grain
unloaded in the first quarter was 175 tons. for  = 0.1 and
0.5. Quarter Actual tonnage Unloaded
1 180
2 168
3 159
4 175
5 190
6 205
7 180
8 182
 Comparison of Forecast Error
Rounded Absolute Rounded Absolute
Actual Forecast Deviation Forecast Deviation
Tonnage with for with for
Quarter Unloaded  = .10  = .10  = .50  = .50

1 180 175 5.00 175 5.00

2 168 175.5 7.50 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 175 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 205 175.02 29.98 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62

46
 Comparison of Forecast Error
∑ |deviations|
Rounded Absolute Rounded Absolute
MADActual
= Forecast Deviation Forecast Deviation
Tonnage n
with for with for
Quarter Unloaded  = .10  = .10  = .50  = .50
1
For 180
= .10 175 5.00 175 5.00
2 168 = 82.45/8
175.5 = 10.31
7.50 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 For 175
= .50 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 205 = 98.62/8
175.02 = 12.33
29.98 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62
 Comparison of Forecast Error
∑ (forecast errors)2
Rounded Absolute Rounded Absolute
MSE = Actual Forecast Deviation Forecast Deviation
Tonnage
n
with for with for
Quarter Unloaded  = .10  = .10  = .50  = .50
1
For 180
= .10 175 5.00 175 5.00
2 = 1,526.54/8
168 175.5 = 190.82
7.50 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 For 175
= .50 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 = 1,561.91/8
205 175.02 = 195.24
29.98 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62
MAD 10.31 12.33
 Comparison of Forecast Error
n
∑100|deviationi|/actuali
Rounded Absolute Rounded Absolute
MAPE = i=1
Actual Forecast Deviation Forecast Deviation
Tonnage withn for with for
Quarter Unloaded  = .10  = .10  = .50  = .50
1
 = .10 175
For 180 5.00 175 5.00
2 168 = 44.75/8
175.5 = 7.50
5.59% 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 =
For 175 .50 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 205 = 54.05/8
175.02 =29.98
6.76% 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62
MAD 10.31 12.33
MSE 190.82 195.24
 Comparison of Forecast Error

Rounded Absolute Rounded Absolute

Actual Forecast Deviation Forecast Deviation
Tonnage with for with for
Quarter Unloaded  = .10  = .10  = .50  = .50
1 180 175 5.00 175 5.00
2 168 175.5 7.50 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 175 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 205 175.02 29.98 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62
MAD 10.31 12.33
MSE 190.82 195.24
MAPE 5.59% 6.76%
 Time Series vs. Causal Models
Time series compares data being forecast over
time, i.e. Time is the independent variable or
x- axis or X-variable.
Causal models compare data being forecast
against some other data set which the forecaster
may think is a cause of the forecasted data, e.g.
population size causes newspaper sales.

11-51
 Causal Forecasting Models

•The general regression model:

𝑦 = 𝑎 + 𝑏𝑥

•The Values of a and b

11-52
 Least Squares Method

Actual observation Deviation7

Values of Dependent Variable

(y-value)

Deviation5 Deviation6

Deviation3

Deviation4

Deviation1
(error) Deviation2
Trend line, y ^= a + bx

Time period
 Least Squares Method

Actual observation Deviation7

Values of Dependent Variable

(y-value)

Deviation5 Deviation6

Deviation3
Least squares method minimizes the sum
of the squared errors (deviations)
Deviation4

Deviation1
(error) Deviation2
Trend line, y ^= a + bx

Time period
 Least Squares Method

Equations to calculate the regression variables

^
y = a + bx

xy - nxy
b=
x2 - nx2

a = y - bx
 Least Squares Example

Time Electrical Power

Year Period (x) Demand (megawatt) x2 xy
2006 1 74 1 74
2007 2 79 4 158
2008 3 80 9 240
2009 4 90 16 360
2010 5 105 25 525
2011 6 142 36 852
2012 7 122 49 854
∑x = 28 ∑y = 692 ∑x2 = 140 ∑xy = 3,063
x=4 y = 98.86

∑xy - nxy 3,063 - (7)(4)(98.86)

b= = = 10.54
∑x - nx
2 2 140 - (7)(42)

a = y - bx = 98.86 - 10.54(4) = 56.70

 Least Squares Example

Time Electrical Power

Year Period (x) Demand x2 xy
2003 1 74 1 74
2004 2 79 4 158
2005The trend
3 line is 80 9 240
2006 4 90 16 360
^
2007 y =5 56.70 + 10.54x
105 25 525
2008 6 142 36 852
2009 7 122 49 854
∑x = 28 ∑y = 692 ∑x2 = 140 ∑xy = 3,063
x=4 y = 98.86

∑xy - nxy 3,063 - (7)(4)(98.86)

b= = = 10.54
∑x - nx
2 2 140 - (7)(4 2)

a = y - bx = 98.86 - 10.54(4) = 56.70

 Least Squares Example

Trend line,
160 –
y^= 56.70 + 10.54x
150 –
140 –
130 –
Power demand

120 –
110 –
100 –
90 –
80 –
70 –
60 –
50 –
| | | | | | | | |
2006 2007 2008 2009 2010 2011 2012 2013 2014
Year
 Example of Time Series Model

Yt = a + b(t)
t Dt Ft
1 120 119.52
2 124 121.18
Dt = actual sales
3 119 122.84
4 124 124.5
Ft = forecasted sales
5 125 126.15
6 130 127.81
t = time period (e.g. year)
7 129.47

Intercept (a) 117.8667

Slope (b) 1.657143

F7 = 117.87 + 1.66 (7) = 129.47 = sales forecast for seventh year

11-59
 Selecting a Forecasting Methods
During selecting forecasting, you should have to:

 Use and decision characteristics

➢ Accuracy required, Time horizon
➢ Pricing decision require highly accurate short ranged forecasts for large
number of item

 Data availability and quality

 Data pattern affects the type of forecasting:

➢ If the time series is flat, A first order method can be used, where as if the
data shows trend or seasonal pattern some advance method will be used.
➢ If the data is unstable over time, a qualitative method may be selected.

 Don’t force the data to fit the model!

60
 Application of forecasting
• Forecasts are vital to every business organization
and for every significant management decision.
➢ Sales Forecasting : Any company in selling
goods needs to forecast the demand for those goods.

➢ Forecasting Economic Trends : forecasting economic

trends on a regional, national, or even international level.
➢ Forecasting Staffing Needs:
➢ Forecasting in education environment :
➢ Ministry of Petroleum :
➢ Department of Technology:
61
62